Dynamic Adaptive Workflow Scheduling for Instance Intensive Cloud Applications
Muthukathan Rajendran Sudha, M. Monica · JES (Jurnal Ekonomi Syariah) · 2012
Abstract - Large scale data intensive business and scientific computations from various scientific endeavors are composed as workflows that access shared data, shared pool of resources on demand. Predicting the dynamic properties of the resources and coordinated control of resources and workflows is critical especially selection of algorithms for scheduling cloud workflows in assorted environments built on different system architectures is the most challenging. Most of the cloud applications are data intensive containing large datasets having huge number of workflow instances running on a cloud computing environment which lead to an ever growing need for optimal scheduling of task among the cloud network devices Also cloud computing technology rely on different system infrastructures in the form of grid structures, or actually distributed services (service-based). In cloud computing infrastructure with a “pay per use” context (business-based) the execution cost should be considered which rely on the actual execution time. Each of these different infrastructures demands various workflows scheduling algorithm. The investigation of cloud workflow systems, is a timely issue and worthwhile for increasing performance efforts. The intent of this research work is to evaluate the limitation of existing algorithms and to work on suitable scheduling mechanisms for cloud workflows.